Abstract — Trace fossils record foraging behaviours, the search for resources in patchy environments, of animals in the rock record.  Quantification of the strength, density and nature of foraging behaviours enables the investigation of how these may have changed through time.  Here, we present a novel approach to explore such patterns using spatial point process analyses to quantify the scale and strength of ichnofossil spatial distributions on horizontal bedding planes.  To demonstrate the utility of this approach we use two samples from the terminal Ediacaran Shibantan Member in South China (between 551 and 543 Ma) and the early Cambrian Nagaur Sandstone in northwestern India (between 539 and 509 Ma).  We find that ichnotaxa on both surfaces exhibited significant non-homogeneous lateral patterns, with distinct levels of heterogeneity exhibited by different types of trace fossils.  In the Shibantan, two ichnotaxa show evidence for mutual positive aggregation over a shared resource, suggesting the ability to focus on optimal resource areas.  Trace fossils from the Nagaur Sandstone exhibit more sophisticated foraging behaviour, with greater niche differentiation.  Critically, mark correlation functions highlight significant spatial autocorrelation of trace fossil orientations, demonstrating the greater ability of these Cambrian tracemakers to focus on optimal patches.  Despite potential limitations, these analyses hint at changes in the development and optimisation of foraging at the Ediacaran–Cambrian transition and highlight the potential of spatial point process analysis to tease apart subtle differences in behaviour in the trace fossil record.

The Shibantan bed-sole surface was chosen because there was a reasonable area of bedding plane with abundant ichnofossils and limited evidence of preservational or erosional heterogeneity (Fig. 4A; see also fig. 6a of Xiao et al. 2021).  Taxonomy of Ediacaran trace fossils can be highly controversial (Darroch et al. 2021), and so here we recognize two distinct morphogroups consisting of relatively large (~1 cm) and small (~1 mm) horizontal traces, with no overlap in width.  The smaller ichnofossils are Helminthoidichnites-like, and the larger ones are characterized by poorly preserved spiral structures reminiscent of Streptichnus (Fig 4E; Jensen and Runnegar 2005; Xiao et al. 2021).  Importantly, even if this preferred taxonomy is incorrect, the statistical analyses outlined below will not be impacted, as these large  trace fossils represent similar behaviours, likely made by the same progenitor.

The Naguar bed-sole surface similarly contains limited preservational or erosional heterogeneity (Fig. 4C).  We mapped four distinct types of trace fossils on this bedding plane. Of primary interest are larger horizontal traces, similar to Treptichnus pedum previously identified from this unit (Pandey et al. 2014), and smaller horizontal trails ~1 mm wide (Fig. 4D, white oval). The larger trails may not contain all the diagnostic features of T. pedum, so we classify them more broadly as Treptichnus isp. These exhibit a probing behaviour broadly similar to that of Streptichnus from the Shibantan. Smaller trails were regarded as Planolites isp. (Pandey et al. 2014). Circular traces are of similar diameter to the width of the larger horizontal traces, and the two are often associated. A fourth group of bilobed traces represent Rusophycus and/or Cruziana (Fig. 4F; Pandey et al. 2014).

The outline of each bed and individual trace fossils were marked as vector lines within Inkscape 0.92.3 (Fig. 5).  Areas with low fossil density, interpreted as artefacts of differing erosion and/or poor photographic contrast, were excluded (Fig. 5B).  Critically, such areas are much larger than the centimetre scale heterogeneities investigated here, and thus would not affect this study.  Most of the traces can be clearly marked, however two different modes were employed for larger horizontal trails (Streptichnus and Treptichnus).  Initially, each individual trace was marked irrespective of adjacent ones (unconnected trails, Fig. 4B, D).  Then, in instances where traces appear to represent a series of segments/steps that can be reliably ascribed to the same progenitor, they were connected or simplified to a single segment to focus on the overall paths of the tracemakers (connected trails, Fig. 4B, D).  While traces are continuous objects, rather than points, these spatial methods have already been established for modern organisms, such as root systems, so are appropriate to apply to trace segments (Kitchell 1979; Koy and Plotnick 2010). Each single line represents a single action of the trace maker so that as long as consistent labelling approaches are applied across each surface, this marking up approach results in a representation of the behaviours of the trace makers across the bedding planes.

Line data was extracted from Inkscape using a custom script written in Haskell (https://github.com/egmitchell/traces).  Analyses were performed in R (R Core Team 2017) using spatstat (Baddeley and Turner 2005).